This study will compare the unconditioned and conditioned effects of placebo capsules in the treatment of patients with borderline hypertension (BH). The use of placebos is routine in trials of antihypertensive medication, but the placebo is nearly always given before the active drug, and most studies have relied on traditional clinic measurement of blood pressure (BP). However, ambulatory BP monitoring (ABPM) is the gold standard for evaluating the effectiveness of any antihypertensive treatment, and placebo effects on ABP have mostly been negligible. However, there is also evidence that when a placebo is given after the active drug, there may be a more profound effect. A long series of both animal and human studies by the co-principal investigator (Dr. Ader) have shown that such learned or classically conditioned placebo effects may have genuinely therapeutic effects. One such study, which will provide the basis for the proposed study, found that placebo given after the active drug lowered home BP, but ABP was not evaluated. The proposed study will extend the previous one, and will be performed in Community Health Centers, in a culturally diverse, economically disadvantaged population. 120 patients with untreated BH will have their BPs monitored by clinic, home, and ABP recording. After a 4-week run-in period of telephone-linked home monitoring (continued throughout the study) they will be randomized to 3 groups: no drug, active drug (a beta blocker-diuretic combination), or matching placebo. This will enable evaluation of the unconditioned placebo effect. For the next 4 weeks all 3 groups will receive active drug. ABPM and clinic BP will be measured at the end of each period. For the final period patients in each group will be randomized to no drug or placebo. ABPM and clinic BP will be measured after 2 weeks, but home BP will continue until BP returns to 140/90 mmHg. It is hypothesized that the conditioned placebo effect will result in a delayed return of BP to pretreatment levels, and the inclusion of the No Drug group will distinguish this from the effects of drug washout. To promote and assess compliance with the medications, electronic containers (MEMscaps) will be used. The demonstration of a conditioned placebo effect that produces a sustained reduction of BP could result in more economical treatment with fewer side effects.